Challenge

This project functions as a submodule within an evolving BMW initiative, designed to integrate seamlessly into a comprehensive database and data streaming infrastructure. The primary challenge  is how to achieve this integration effectively, given the limited direct access to real B2G users—particularly in terms of data streaming technology—while maintaining compatibility with the BMW ecosystem.

Approach

In light of these challenges, I structured the design and frontend specifications into four distinct phases, maintaining a strong emphasis on user-centered principles. This approach was especially important in a complex project where direct access to real users was limited.

Workshop

In our initial four-hour workshop with stakeholders and developers, we used collaborative techniques to identify challenges, define key audiences, and align with corporate goals. Through agile methods and a focus on actionable insights, we created a strategic roadmap tailored to the project’s time and budget constraints. I decided to use the SCRUM framework with tailored adjustments and extensions in designing and specifying frontend.

Focus group discussion & stakeholders interview

In the second four-hour workshop, I led in-depth research and discussions with stakeholders and the customer-facing team to define user needs, business goals, and technical requirements. This collaborative session provided valuable insights from multiple perspectives, giving us a clear understanding of the project’s objectives and constraints. This session revealed several key findings.

Gathering information

During this workshop, we utilized "Post-it" notes and the "Affinity Diagramming" technique to organize and prioritize key insights. This approach helped us identify core requirements and categorize relevant information.

Content audit

Following feedback from a workshop with the customer-facing team and law enforcement agencies, we performed a content audit to catalog the platform’s initial information and features.

Adjustment

We acknowledged that data retrieval solutions might not suit all users or particular use cases. Specifically, API solutions may fall short when it comes to accessing historical data, which poses challenges for certain requirements. To address this gap, we have actively continued conducting another workshop aimed at identifying a third solution tailored for these special cases.

Key areas of focus 

We focused on two main user groups—law enforcement agencies and BMW’s internal team

Automated platform for law enforcement:

• Enables VIN-based data requests for multiple data types using various retrieval methods.

• Supports data extraction via APIs, downloadable files, or streaming.

• Facilitates vehicle information updates and tracking based on state models.

Automated Platform for BMW Internal Teams:

• Provides verification, approval, and auditing of vehicle data requests to ensure regulatory compliance.

• Allows the creation and provisioning of data packages across different retrieval methods.

• Supports vehicle information updates and state-model-based tracking.

• Ensures high data security and privacy for all transactions and records.

Seamless BMW Ecosystem Integration:

• The platform integrates smoothly with BMW’s existing systems and infrastructure for efficient, cohesive operations.

Building the site structure

Everything has finally come together! I’m now developing high-fidelity prototypes for each requirement and conducting direct testing with stakeholders. This approach enables us to gather immediate feedback and refine the designs before moving on to user stories.

Example

One key requirement I managed was allowing B2G users to request both time-sensitive and time-insensitive data for a vehicle in a single session. I gathered requirements in a stakeholder workshop, prioritized the requirement, coordinated between stakeholders and developers, designed and tested  a solution with a popup form that enables B2G users to customize their data extraction preferences. Users can choose specific data types for extraction, with at least one time period required for time-sensitive data. The interface also allows users to request data across multiple time periods and to adjust selections by removing any added time periods or data types as needed. Once configured, the selected data is displayed in a table, automatically organized into time-sensitive and time-insensitive categories for clarity and easy access.

Iterative user testing

We structured feature testing into three phases: technical testing, usability testing, and user acceptance testing. As the UX Designer, I led bi-weekly usability testing based on the MVP. Given limited access to actual users for this B2G product, we used a range of alternative approaches:

Proxy users from customer-facing teams: Team members with customer interactions provided valuable feedback during usability tests.

Involvement of stakeholders and domain experts: Stakeholders and experts contributed through heuristic evaluations, offering insights from their areas of expertise.

Simulated user scenarios with testing team: We created detailed, realistic scenarios reflecting user workflows and conducted pilot tests with team members unfamiliar with the design to uncover usability issues.

Other projects